IPO Market 2026: The First Half in Numbers

The first half of 2026 put 184 new US-dollar listings on American exchanges, raising 133.5 billion dollars, and a single June deal accounts for 56.2 percent of that money, more than every other listing combined. This page is the half's listing ledger: how many deals priced and when, what listed and where, what a typical deal raised, how the pace compares with every first half since 2019, and how the class trades against its offer prices.

How many IPOs priced in the first half of 2026?

First, what is being counted. The universe here is completed listings on US exchanges, priced in US dollars, from offering records, every listing type included. That is broader than the headline "IPO" of financial television: alongside operating-company debuts, it counts closed-end funds, trusts, and blank-check vehicles. Narrower definitions produce smaller counts, which is why published "how many IPOs" figures disagree, the answer depends on the fence around the word. The fence used here is stated in the SQL under every panel, and the mix inside it gets its own panel below. The half, month by month:

SELECT toStartOfMonth(listing_date) AS month,
       count() AS listings,
       round(sum(total_offer_size) / 1e9, 1) AS raised_bn_usd
FROM global_markets.stocks_ipos
WHERE ipo_status = 'history'
  AND currency_code = 'USD'
  AND listing_date >= '2026-01-01'
  AND listing_date <= '2026-06-30'
GROUP BY month
ORDER BY month

The cadence was lumpy: 45 listings in February against 15 in March, the half's quietest month. The money was lumpier still, June's 86.8 billion towers over every other month (May's 16.2 billion is the runner-up), and one deal is most of the reason.

The half in one row

Totals, the typical deal, and the two concentration receipts:

SELECT count() AS listings,
       round(sum(total_offer_size) / 1e9, 1) AS raised_b,
       round(sumIf(total_offer_size, listing_date >= '2026-06-01') / 1e9, 1) AS june_raised_b,
       round(100 * toFloat64(sumIf(total_offer_size, listing_date >= '2026-06-01')) / toFloat64(sum(total_offer_size)), 1) AS june_share_pct,
       round(max(total_offer_size) / 1e9, 1) AS biggest_deal_b,
       round(100 * toFloat64(max(total_offer_size)) / toFloat64(sum(total_offer_size)), 1) AS biggest_deal_share_pct,
       countIf(total_offer_size >= 1000000000) AS billion_dollar_deals,
       round(avg(total_offer_size) / 1e6, 0) AS average_deal_m,
       round(quantileExact(0.5)(toFloat64(total_offer_size)) / 1e6, 0) AS median_deal_m,
       countIf(final_issue_price > 0) AS with_offer_price,
       (SELECT uniqExact(toDate(toTimeZone(window_start, 'America/New_York')))
        FROM global_markets.delayed_stocks_minute_aggs
        WHERE ticker = 'SPY'
          AND window_start >= toDateTime('2026-01-01 00:00:00')
          AND window_start < toDateTime('2026-07-01 00:00:00')) AS trading_sessions
FROM global_markets.stocks_ipos
WHERE ipo_status = 'history'
  AND currency_code = 'USD'
  AND listing_date >= '2026-01-01'
  AND listing_date <= '2026-06-30'

184 listings across 123 trading sessions, more than one new listing per session. 11 deals raised a billion dollars or more, topped by a single largest deal of 75 billion dollars. The concentration columns are the half's defining numbers: June carried 65 percent of the dollars, and the one biggest deal carried 56.2 percent, a majority of everything raised, by one company.

The typical deal lives in a different universe: the average listing raised 738 million dollars, the median, the deal in the middle of the stack, 200 million. A gap that wide means skew: one giant deal drags the mean while the middle of the market prices ordinary-sized offerings. The median is the honest measure of a typical deal.

What kind of companies went public?

The intro's caveat, that the count blends listing types, gets its own panel here.

SELECT multiIf(security_type IN ('SP', 'UNIT'), 'Unit offerings (share + warrant or right)',
               security_type = 'ADRC', 'American depositary receipts (ADRs)',
               'Common or ordinary shares') AS listing_type,
       count() AS listings,
       round(sum(total_offer_size) / 1e9, 1) AS raised_bn_usd,
       round(quantileExact(0.5)(toFloat64(total_offer_size)) / 1e6, 0) AS median_deal_m,
       count() - countIf(match(security_description, '(?i)war|rts')) AS without_warrant_or_right
FROM global_markets.stocks_ipos
WHERE ipo_status = 'history'
  AND currency_code = 'USD'
  AND listing_date >= '2026-01-01'
  AND listing_date <= '2026-06-30'
GROUP BY listing_type
ORDER BY listings DESC

A unit offering bundles one share with a fraction of a warrant or a right, the structure blank-check companies (SPACs) use to come public and hunt for a merger target. Units were the half's biggest bucket by count: 101 listings, all but 1 carrying a warrant or right, raising 18.2 billion dollars, big in count, small in money. The 78 common- and ordinary-share listings raised 114.2 billion, the overwhelming bulk of the half's dollars. The remaining 5 were American depositary receipts (ADRs), foreign companies listing in the US through bank-issued receipts, at 1.2 billion.

One honesty note: the record separates structure, not business model, a closed-end fund lists ordinary shares the same way a rocket company does, so the common-share bucket still blends operating companies with listed funds.

The deals that dominated

The top of the table is where the half's money actually lives:

SELECT issuer_name AS company,
       toString(listing_date) AS listed,
       round(total_offer_size / 1e9, 2) AS raised_bn_usd
FROM global_markets.stocks_ipos
WHERE ipo_status = 'history'
  AND currency_code = 'USD'
  AND listing_date >= '2026-01-01'
  AND listing_date <= '2026-06-30'
  AND total_offer_size > 0
ORDER BY total_offer_size DESC
LIMIT 8

At the top sits Space Exploration Technologies Corp., listed 2026-06-12, at 75 billion dollars, dwarfing every other deal in the table, the listing event of the half by any measure. We covered that debut in detail: SpaceX's first month of trading and its fast-follow Nasdaq-100 index add. Below it, the billion-dollar tier, Cerebras Systems Inc (5.55 billion) and peers, reads like the year's private-market greatest hits going public.

A note on reading offer sizes: the figure is the money raised in the offering, not the company's valuation. A company can float a small slice of itself at a giant valuation, or a big slice at a modest one, what is a stock float measures exactly that gap. Offer size measures the event, not the enterprise.

Nasdaq or NYSE: where the half listed

The venue split, count, dollars, and how much of each venue's count was blank-check supply:

SELECT multiIf(primary_exchange = 'XNAS', 'Nasdaq',
               primary_exchange = 'XNYS', 'NYSE',
               primary_exchange = 'XASE', 'NYSE American',
               'OTC markets') AS exchange,
       count() AS listings,
       round(sum(total_offer_size) / 1e9, 1) AS raised_bn_usd,
       countIf(security_type IN ('SP', 'UNIT')) AS unit_offerings
FROM global_markets.stocks_ipos
WHERE ipo_status = 'history'
  AND currency_code = 'USD'
  AND listing_date >= '2026-01-01'
  AND listing_date <= '2026-06-30'
GROUP BY exchange
ORDER BY listings DESC

Nasdaq took 130 of the half's listings and 116.7 billion of its dollars, a column that includes the June mega-deal, which listed there. NYSE took 44 listings and 16.4 billion. The unit column explains part of the gap: 77 of the leader's listings were blank-check units, its long-standing bread and butter at the small end of the calendar. NYSE American, the small-cap venue, hosted 9 listings for well under a billion combined.

Is 2026 a busy IPO year?

The usual answer lays a half-year count against full-year history, apples to oranges. The panel below fixes the fence: January-through-June listings only, every year since 2019, computed identically.

SELECT toYear(listing_date) AS year,
       count() AS listings,
       round(sum(total_offer_size) / 1e9, 1) AS raised_bn_usd
FROM global_markets.stocks_ipos
WHERE ipo_status = 'history'
  AND currency_code = 'USD'
  AND listing_date >= '2019-01-01'
  AND listing_date <= '2026-06-30'
  AND toMonth(listing_date) <= 6
GROUP BY year
ORDER BY year

184 listings make this the busiest first half since 2021, above 2025's 177 and 2024's 153, the third straight yearly rise from 2023's 125. The 2021 peak of 542, when blank-check listings turned the calendar into a conveyor belt, remains in a different universe. The dollar column is stranger: 133.5 billion is the second-biggest first-half raise in the table, behind only 2021's 167.7, with the standing caveat that one deal is most of the distance between this half and an ordinary one.

One caution: doubling a half-year count assumes the second half looks like the first, and listing activity is famously windowed, the February-to-March swing in the monthly table shows how fast the window closes. Treat the half figure as a measurement, and any full-year projection built on it as a guess.

How has the class of 2026 traded since listing?

Counting deals measures supply; the question a reader actually brings to "IPO market 2026" is whether these listings made money. The panel takes every H1 listing with a recorded final offer price (178 of the 184 carry one) and compares its last regular-hours close of the second half's first full week (July 6-10) with what offering buyers paid.

SELECT multiIf(i.security_type IN ('SP', 'UNIT'), 'Unit offerings',
               i.security_type = 'ADRC', 'ADRs',
               'Common or ordinary shares') AS listing_type,
       count() AS deals_with_offer_price,
       countIf(p.last_close > 0) AS traded_jul6_10,
       countIf(p.last_close > toFloat64(i.final_issue_price)) AS above_offer,
       countIf(p.last_close > 0 AND p.last_close < toFloat64(i.final_issue_price)) AS below_offer,
       countIf(p.last_close = toFloat64(i.final_issue_price)) AS at_offer,
       round(quantileExactIf(0.5)((p.last_close / toFloat64(i.final_issue_price) - 1) * 100, p.last_close > 0), 1) AS median_return_pct
FROM global_markets.stocks_ipos AS i
LEFT JOIN (
    SELECT ticker, toFloat64(argMax(close, window_start)) AS last_close
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE window_start >= toDateTime('2026-07-06 00:00:00')
      AND window_start < toDateTime('2026-07-11 00:00:00')
      AND (toHour(window_start) * 60 + toMinute(window_start)) BETWEEN 810 AND 1199
    GROUP BY ticker
) AS p ON p.ticker = i.ticker
WHERE i.ipo_status = 'history'
  AND i.currency_code = 'USD'
  AND i.listing_date >= '2026-01-01'
  AND i.listing_date <= '2026-06-30'
  AND i.final_issue_price > 0
GROUP BY listing_type
HAVING countIf(p.last_close > 0) > 0
ORDER BY deals_with_offer_price DESC

Three different markets live in that table. The 70 common- and ordinary-share listings that traded during the week split 45 above their offer price against 25 below, with a median gain of 3.6 percent, the typical traditional listing modestly up, a large minority underwater. The unit offerings barely moved: a median of 1.1 percent above offer, the blank-check structure behaving as designed, a unit is mostly a claim on a trust account holding the offering cash, and it tends to sit near the offer price until a merger target appears. The ADR bucket is the half's one clean downside pocket: every ADR that printed a trade that week sat below its offer price, a median of -17.4 percent under, 5 listings, a sample worth exactly that much weight.

Coverage is disclosed in the table: only 56 of 101 unit tickers printed a regular-hours trade that week, unit symbols are thin, and a unit can be split into share and warrant components that trade separately, so unit coverage is inherently partial.

What this means if you're weighing new listings

Three observations travel beyond this half. First, headline counts overstate how many businesses actually arrived: the biggest bucket by count was blank-check units, so a raw "IPO count" mostly measures shell supply. Second, dollar totals hinge on single decisions: one deal carried the majority of the half's raise, so a year-over-year "IPO proceeds" comparison can swing on one boardroom. Third, the aftermarket is split, not uniformly hot: 25 of the 70 traditional listings that traded sat below offer, dispersion, not a rising tide.

Two mechanics are worth knowing before touching any new listing. Early trading runs on a thin float: early prices are jumpier than the same stock will be a year later. And insider lock-ups conventionally expire about 180 days after listing, which puts this class's lock-up expirations on the second half's calendar, a scheduled step-up in sellable shares. Neither point says buy or avoid; both say the offer-price return above is not what a market buyer could capture, offer allocations go mostly to institutions; everyone else buys the open.

The second half is already writing its own ledger, the first full July week delivered a mega-listing from a Korean memory-chip maker, but that raise belongs to the H2 edition. The market these deals priced into is measured in the H1 2026 market recap.

IPO market 2026 FAQ

How many IPOs have there been in 2026 so far?

184 completed US-dollar listings priced in the first half of 2026, counting every listing type: 78 common- or ordinary-share listings, the traditional "IPO", and 101 blank-check unit offerings among them.

What was the biggest IPO of 2026 so far?

Space Exploration Technologies Corp., listed 2026-06-12, raised 75 billion dollars, the largest US listing of the half by a wide margin, and 56.2 percent of the half's entire raise by itself.

How many 2026 IPOs were SPACs?

101 of the half's 184 listings were unit offerings, the share-plus-warrant-or-right bundle blank-check companies use, raising 18.2 billion dollars: a large share of the count, a small share of the money.

How have 2026 IPOs performed since listing?

At the last regular-hours close of the July 6-10 week, the median common- or ordinary-share listing traded 3.6 percent above its offer price, 45 above, 25 below. Unit offerings clustered just above their offer price; every ADR that traded sat below its offer.

Is the IPO market hot right now?

By count, H1 2026 was the busiest first half since 2021, 184 listings against 2025's 177, like-for-like, though far under 2021's 542. By dollars it ranks second only to 2021, a rank that rests largely on one mega-deal atop an otherwise ordinary market.

Every figure above is a stored, versioned query over listing records, expand the SQL under any panel, or slice the IPO calendar your own way on the Strasmore terminal.